王雪亮+谭柱+王小丽+刘晖

摘要：基于桅杆结构拉耳表面应变对其焊缝裂纹扩展长度的敏感度分析，提出了桅杆结构拉耳焊缝损伤的两级模糊模式识别方法，并通过试验研究进行验证。采用有限元模拟建立了拉耳索力-关键点应变-焊缝裂纹扩展长度的模糊模式库；对16个不同裂纹长度的拉耳模型进行试验，以实测拉耳表面关键点应变为待识别样本，采用模糊模式识别方法对索力及焊缝裂纹扩展长度进行分步识别。结果表明：模糊数据库中的模拟数据能够反映拉耳的实际受力状态，识别焊缝裂纹扩展长度与实际裂纹长度具有较好的吻合度；对于不同索力及裂纹长度，识别焊缝裂纹扩展长度的误差在1 mm之内，最大隶属度均大于0.8，识别峰值明显；该识别技术具有可行性，是实现桅杆结构损伤智能识别的一种有效手段。

关键词：桅杆结构；焊缝裂纹扩展长度；模糊模式识别；拉耳应变

中图分类号：TU347 文献标志码：A

文章编号：1673-2049（2017）02-0079-07

Abstract：Base on sensitivity analysis of surface strain of ear-plate connection in guyed mast on crack propagation length， a two level fuzzy pattern recognition method was proposed for the damage of welded ear-plate connection in guyed mast， and was verified by experimental study. The fuzzy pattern database of cable forces and key-point strains of ear-plate connections， as well as weld crack propagation lengths was developed by finite element. 16 ear-plate connection models with different crack lengths were tested， and the key-point strains were measured to identify cable forces and weld crack propagation lengths by the proposed method. The results show that the simulated data in fuzzy database can reflect the actual stress state of ear-plate connection， and the identified weld crack lengths agree well with the actual weld crack lengths. For the different cable forces and crack propagation lengths， the error of identified weld crack length is within 1 mm， the maximum membership degree is greater than 0.8， and the recognition peak is obvious. The recognition technology is feasible and is an effective method to realize the damage intelligent identification of guyed mast.

Key words：guyed mast； weld crack propagation length； fuzzy pattern recognition； strain on ear-plate connection

0引 言

桅杆结构由于质量轻、布置灵活等特性在广播电视、通信气象、航天航空等领域得到广泛应用，但国内外桅杆结构的倒塌灾害频繁发生，对生产生活造成极大损失。这些灾害中绝大多数桅杆结构的破坏发生在风荷载不大的情况下，究其原因主要是风荷载作用下结构疲劳所致[1-3]，而拉耳连接焊缝的疲劳损伤是重要因素之一。由于纤绳和桅杆的耦合作用导致在随机风荷载作用下产生复杂振动，如横风振動、侧向驰振等，这种长期振动造成了构件和拉耳连接的疲劳损伤，从而导致结构破坏。……